Is prostate-specific antigen density more useful than prostate-specific antigen levels in the diagnosis of prostate cancer?

OBJECTIVES: To compare the performance of prostate-specific antigen (PSA) levels with the performance of PSA density (PSAD), the ratio of the serum to the size of the prostate, as predictors of the presence of prostate cancer.
METHODS: We analyzed the results of digital rectal examination (DRE), transrectal ultrasonography (TRUS), serum PSA levels, and PSAD in 244 patients who had a needle biopsy of the prostate.
RESULTS: Cancer was detected in 110 patients (45%). Compared with DRE, TRUS and serum PSA levels 4.0 ng/mL or higher, PSAD at a cutoff point of 0.15 ng/mL/cm3 was significantly more specific and had a higher positive predictive value than each of the other tests but was significantly less sensitive than TRUS and PSA (P < 0.05 for each). In a receiver operating characteristic analysis, PSAD was significantly more accurate than PSA (P < 0.001). In 80 patients with a normal PSA, PSAD added no additional information, and PSAD was not able to identify a subset at low risk. In 82 patients with a high PSA level (10 ng/mL or higher), 15% had a PSAD less than 0.15 and only 8% had a cancer.
CONCLUSIONS: Overall, PSAD was significantly more accurate than PSA for predicting the results of needle biopsy of the prostate, but in practice PSAD proved useful in only a small subset of patients. If the serum PSA level was high but the PSAD was low, cancer was rarely detected. These patients may be suitable candidates for careful follow-up rather than early repeat biopsy.